DESCRIPTION: Maximum life spans are significantly increased in mice by Leg 1a, a genetic variant in Mus m molossinus (MOLD) on chromosome 8 near microsatellite marker D8Mit171. The same chromosome region reduces blood insulin without affecting glucose levels or normal growth. Defining this gene that retards mammalian aging may suggest clinical treatments to improve health in old age. Aim 1 tests if Leg1a has effects on aging like those already defined for caloric restriction (CR) by comparing four groups of 84 mice: Leg1a carriers and non-carriers, each with a fed ad lib group and a CR group. At 4 and 24 months of age, biological aging will be assessed with a battery of non-harmful measures including blood insulin and glucose, T cell subset distributions, wound healing and collagen denaturation. In addition, growth will be followed from weaning, and life spans and pathology will be determined. Splenic T cell telomere lengths will be tested in separate groups. Aim 2 maps Leg1a to a 1 cM region, and tests if Leg1 is the locus that regulates insulin. Progeny will be produced from 10 males; each with a different chromosomal recombination carrying a unique subset of MOLD alleles in the 10 cM region now defined. For statistical power, each male will produce 84 progeny carrying the recombination and 84 non-carrier controls with no MOLD alleles in the 10 cM region. Insulin and glucose levels will be tested at 4 months in these mice, and life spans of the longest-lived 18 and 40 percent will be compared to map Leg1a to a 1 cM region. If the long-lived recombinant groups all have low insulin, it will strongly support the hypothesis that Leg1a regulates insulin. Aim 3 uses the progeny of Aim 2 to map the locus regulating insulin to within 1 cM. Subsequently, progeny from 10 males with recombinations within this 1 cM region will be used to refine the map location to 0.1 cM (roughly 10 genes), comparing 79 carriers of each recombination with 79 non-carrier progeny. Candidate genes will be tested to see whether MOLD alleles differ from those of NZW, LP and BALB. The gene will ultimately be identified by homologous recombination.